The heating, ventilation, and air-conditioning (HVAC) systems are one of the main factors that contribute to the building’s energy usage. Achieving an effective balance between reducing energy use and maintaining acceptable thermal comfort is the key challenge in conventional HVAC systems. To overcome this challenge, integrating the occupant-centric controls coupled with digital twins into HVAC systems is another potential technique for this effective balance. For this purpose, computational fluid dynamics (CFD) offers the potential, in combination with other surrogate models for real- time applications to enhance the system's performance further. In general, the CFD is applied to investigate indoor airflow/temperature distributions. These are essential for occupant health, comfort, and energy optimisation for the HVAC design state. The objective of this study is to propose an initial step toward building an occupant-centric HVAC digital twin by validating a CFD model of an office against dense in-situ sensing data. The model has been used to resolve airflow and temperature stratification under conventional HVAC operations, using ANSYS Fluent. The boundary conditions have been derived from measured supply parameters, internal gains, and local weather conditions. The results from this study show that the air velocity and temperature at selected durations follow the same trend with low errors, compared to the sensing and measurement data. The model validation from this study establishes the basis for a weather- aware, occupant-feedback digital twin for larger floorplates and multi-zone systems. To achieve the target of the energy and comfort co-optimisation in Industry 4.0-ready buildings, the future work will focus on surrogate modelling to enable near-real-time inference for closed-loop occupant-centric controls, which will directly support dynamic set-point adjustments and multi-zone system ventilation.